Research in Developmental Disabilities 35 (2014) 581–590
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Reliability and validity of the Korean version of the Trunk Control Measurement Scale (TCMS-K) for children with cerebral palsy Jun-Young Jeon a, Won-Seob Shin a,b,* a b
Department of Physical Therapy, Graduate School of Daejeon University, Daejeon, Republic of Korea Department of Physical Therapy, College of Natural Science, Daejeon University, Daejeon, Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Article history: Received 3 September 2013 Received in revised form 4 January 2014 Accepted 6 January 2014 Available online 27 January 2014
The Trunk Control Measurement Scale (TCMS) was developed by Lieve Heyrman in 2011 to evaluate the clinical features of impaired trunk control ability in patients with cerebral palsy (CP). This study aimed to demonstrate the reliability and validity of the Korean version of the Trunk Control Measurement Scale (TCMS-K) for children with CP. Fifty children with spastic CP (mean age 9.08 3.75) participated in the study. They were classified using the Gross Motor Function Classification System and the Manual Ability Classification System. The intraclass correlation coefficient (ICC) value of the inter-rater reliability for the TCMS-K was .987–.998, and the intra-rater reliability was .947–.996. The Spearman rank correlation coefficient between the TCMS-K and the Gross Motor Function Measure-B dimension was .860. The results of the study support that the TCMS-K has a high reliability and validity, which is similar to the original version. Thus, the TCMS-K is a suitable evaluation tool to assess the qualitative performance of trunk control and sitting balance for children with CP, and we expect that it will be a very useful tool for clinicians and researchers. ß 2014 Elsevier Ltd. All rights reserved.
Keywords: Cerebral palsy Korean version Trunk Control Measurement Scale Reliability Validity
1. Introduction Cerebral palsy (CP) is a non-progressive and permanent motor and sensory disorder resulting from a permanent defect or lesion in the immature brain. It was increased the incidence of cerebral palsy because of the advances in medical technology and increased use of incubators for baby of premature and low birth weight (Molnar & Alexander, 2006). According to the data of The Korean National Statistical Office (2011), the rate of low birth weight and preterm baby from the total live births was 4.6% and 5.2%, respectively. Hong (2004) reported that 716 children of cerebral palsy caused the low birth weight 57.6%, the premature baby 62.9% in Korea. The high prevalence of cerebral palsy caused by extreme prematurity and low birth weight baby of prior to 32 weeks (Schendel, Schuchat, & Thorsen, 2002). People with CP have significant limitations in activities of daily living due to their lack of motor control ability and balance, the presence of primitive reflexes, spasticity, and voluntary movement difficulties according to the extent of brain impairment (Harbourne, Willett, Kyvelidou, Deffeyes, & Stergiou, 2010). The main problems experienced by children with CP are abnormal posture and movement strategies because of their trunk instability, motor and sensory impairment, joint contracture, muscle shortening, and deformed musculoskeletal system with reduced bone mineral density (Goodman &
* Corresponding author at: Department of Physical Therapy, Daejeon University, 62 Daehak-ro, Dong-gu, Daejeon 300-716, Republic of Korea. Tel.: +82 42 280 2294; fax: +82 42 280 2295. E-mail addresses: [email protected], [email protected] (W.-S. Shin). 0891-4222/$ – see front matter ß 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ridd.2014.01.009
582
J.-Y. Jeon, W.-S. Shin / Research in Developmental Disabilities 35 (2014) 581–590
Fuller, 2008). It has been the primary problem in assessing the status of the children (Brogren, Hadders-Algra, & Forssberg, 1998). Nevertheless, assessment and research on children with CP have focused on upper and lower extremities rather than the trunk, and there is a lack of studies dealing with the examination of trunk control measurement (Damiano, Alter, & Chambers, 2009; Klingels et al., 2010). It has being applied in order to solve these problems of cerebral palsy in Korea that invasive methods include the surgical intervention, the Botox injection, and non-invasive methods include the neurodevelopmental treatment, the Bobath approach, the Vojta approach, proprioceptive neuromuscular facilitation, functional electrical stimulation, biofeedback training, postural control. Generally, trunk control while in a sitting posture develops from two to nine months of age, leading to independent head control for typically developing children (Harbourne & Stergiou, 2003). However, children with neurological damage have a delayed trunk control ability and sitting alignment depending on their degree of impairment (Brogren, Forssberg, & HaddersAlgra, 2001; Liao, Yang, Hsu, Chan, & Wei, 2003). Almost all children with CP have difficulty standing independently and have considerable difficulty with and fear of maintaining a standing position, due to the narrow base of support it provides. Therefore, this posture is not suitable to assess their trunk control ability, but the sitting posture can give them freedom in their daily lives due to the wide base of support and stability it provides. It is the best suitable posture to assess their functional level and balance ability, as it is their main posture for activities of daily living (Brogren et al., 1998). There are some clinical outcome measurements to evaluate the gross motor, function and balance for children of cerebral palsy in Korea that the gross motor function measurement, Gross Motor Function Classification System, pediatric balance scale, etc. But, these tools are limited that evaluate the quality aspect of action and compensation movement occurred during the trunk motion on sitting. The Trunk Control Measurement Scale (TCMS) was developed by Lieve Heyrman in 2011 to assess children with CP regarding their clinical features related to trunk control impairment. It is based on the Trunk Impairment Scale (TIS), the tool used to assess trunk control in adults who have had a stroke (Heyrman et al., 2011). It can be scored the quality and compensation movement of trunk motion. To date, no attempts have been made to develop a Korean version of the TCMS. A Korean version of the scale should have the same meaning and validity as the original version to reduce the confusion with English expressions and cultural differences when translated into Korean (Beaton, Bombardier, Guillemin, & Ferraz, 2000). Using a common questionnaire to objectively assess in another country’s language enables countries and their populations’ unique characteristics to be compared to overcome culture and language differences (Bicer, Yazici, Camdeviren, & Erdogan, 2004). Also, the exchange of information and the development of research can be achieved (Pietrobon, Coeytaux, Carey, Richardson, & DeVellis, 2002). The TCMS should particularly be used to translate the Korean version considering the cultural features of Koreans’ sedentary lifestyle (Bae, Jang, Lee, & Kim, 2012). Thus, the purpose of this study was to determine the reliability and validity of the Korean version of the original TCMS developed in English for children with CP. 2. Materials and methods 2.1. Participants This study was participated 50 children (22 males and 28 females) with spastic CP with mean age of 9.08 years from two rehabilitation clinics in Daejeon city. We included participants who able to sit independent or need to only hand and feet support for at 30 min. They could communicate and understand the instructions for the test. We excluded the child who injected the botulinum toxin or performed orthopedic surgery and selective posterior rhizotomy in the last six months, have the problem of visual and equilibrium system. We also excluded other type of CP (Heyrman et al., 2011). We used the Gross Motor Function Classification System (GMFCS) and Manual Ability Classification System (MACS) to classify the functional level of subjects. Ethical approval was gained by the institutional review board of the Daejeon University. We provided the sufficient explanation about research to children and parents. They were asked to sign a written statement in which they formally consented to the inclusion of the study. Their characteristics are presented in Table 1. 2.2. Measurement tools 2.2.1. Trunk Control Measurement Scale: TCMS The TCMS measures main two components of trunk control on sitting that static and dynamic sitting balance. The dynamic sitting balance is divided into the selective movement control and dynamic reaching. The static sitting balance examines static trunk control during the stable movement of upper and lower limbs on sit posture. The selective movement control of dynamic sitting balance evaluates selected movements of trunk within the base of support in three planes (flexion, extension, lateral flexion and rotation). The dynamic reaching of dynamic sitting balance measures the performance of reaching task during the active trunk movement in three plane while maintaining the balance between the beyond the base of support and the center of gravity. The total scale consist the 15 item, included the five, seven and three items of the each subscales. All items are scored the ordinal scale on two, three or four point, and bilaterally performed if related clinical activities. The total score of the TCMS are from 0 to 58 point, and it have the high score the better performance as GMFCS level is lowered. The reliability for the TCMS was the intraclass correlation coefficient (ICC) .98 for inter-rater, ICC .97 test–retest and the Spearman rank correlation coefficient for construct validity was .6–.87 between TCMS and Gross Motor Function Measure
J.-Y. Jeon, W.-S. Shin / Research in Developmental Disabilities 35 (2014) 581–590
583
Table 1 General characteristics of the subjects (N = 50). Variable Age (years) Sex Male Female Types of cerebral palsy Spastic diplegia Secondary quadriplegia Left hemiplegia Right hemiplegia GMFCS I II III IV V MACS I II III IV V Total
N (%) 9.08 3.75 22 (44%) 28 (56%) 35 6 5 4
(70%) (12%) (10%) (8%)
12 9 10 19 0
(24%) (18%) (20%) (38%) (0%)
15 15 18 2 0 50
(30%) (30%) (36%) (4%) (0%) (100%)
GMFCS, Gross Motor Function Classification System; MACS, Manual Ability Classification System.
(GMFM), excluded A dimension. Thus, the original TCMS has been demonstrated the reliability and construct validity (Heyrman et al., 2013, 2011). 2.2.2. Gross Motor Function Measure: GMFM The GMFM is used to the most common and standardized assessment tool to examine gross motor functional ability for CP. It has been demonstrated the reliability and validity to evaluates the actually possible functional ability and the degree of enhanced movement skill for children (Palisano et al., 2000). It has five dimensions: (A) lying and rolling, (B) sitting, (C) crawling and kneeling, (D) standing, (E) walking, running and jumping. Each item is scored from zero to three, the lower of the GMFM score means the lower of the gross motor function ability. The inter-rater ICC of the GMFM is .89 and the intrarater ICC is .99 (Russell et al., 1989). This study evaluated the only dimension B of the sitting area in order to be compared with the TCMS-K. Dimension A and B could be more easily performed rather than other dimensions for children with CP. Especially, the dimension B was the more relevant with the TCMS than the dimension A in the previous study (Heyrman et al., 2011). Other dimensions (C–E) that included kneeling, standing, walking, running and jumping are considered that limited to compare the balance on sitting because too difficult or impossible to perform for the GMFCS levels II–IV in children with CP. 2.3. Translation procedure The translation procedure of TCMS was consisted of three translation stages included the first and the second translation and the expert committee according to guideline (Beaton et al., 2000). The first translation was achieved using three translators. They were native speakers of Korean, included the two translators who were medical experts and the remaining one who had no medical/clinical background. Each translator translated from original English version to Korean and then made one translation version throughout the discussion. The second translation was achieved by two translators, who were bilingual in English and Korean with no medical knowledge. They translated the result of first stage back into English individually. Finally, the expert committee made a final version of TCMS-K after comparing and analyzing all versions from each stage. The expert committee was consisted of translators participated in the first stage, language professionals, researchers had an experience of questionnaire development and evaluation (Beaton et al., 2000). 2.4. Procedure 2.4.1. Inter-rater and intra-rater reliability Two physical therapists assessed for all subjects with the TCMS-K to examine the inter-rater reliability, the same therapist who conducted the first assessment performed second assessment for all subjects after seven days to investigate the intra-rater reliability.
584
J.-Y. Jeon, W.-S. Shin / Research in Developmental Disabilities 35 (2014) 581–590
2.4.2. Construct validity To determine construct validity of TCMS-K, each therapist examined the GMFM-B dimension and calculated the correlation between the TCMS-K and the GMFM-B dimension (Mawdsley, Hoy, & Erwin, 2000). 2.5. Statistical analysis All data were analyzed using SPSS software version 18.0. Inter-rater and intra-rater reliability for the TCMS-K score was presented by intraclass correlation coefficient (ICC) and the 95% confidence intervals (CI). To investigate construct validity, TCMS-K was compared with GMFM-B dimension score by Spearman rank correlation coefficient. 3. Results 3.1. Inter-rater and intra-rater reliability for the TCMS-K The ICC value of the inter-rater reliability was .987–.998 and the intra-rater reliability was .947–.996. ICC and 95% CI of the subscales and total TCMS-K values are presented in Table 2. 3.2. Construct validity of the TCMS-K The Spearman rank correlation coefficient between the TCMS-K and the GMFM-B dimension was .860 that was significantly positive. 4. Discussion This study was conducted to determine the inter- and the intra-rater reliability and construct validity of the Korean version of the TCMS, the TCMS-K. We introduce a new tool to evaluate the sitting balance and trunk control ability for children with CP. So far, several assessment tools have been developed and used to better understand the examination of motor functions, management protocol decisions, and treatment goal setting for children with CP conducted by many clinicians and researchers (Rosenbaum et al., 2007). For example, the Gross Motor Function Measure (GMFM), Gross Motor Function Classification System (GMFCS), and Manual Ability Classification System (MACS) have demonstrated both high reliability and validity and have been widely used throughout the world (Eliasson et al., 2006; Palisano, Cameron, Rosenbaum, Walter, & Russell, 2006). These tools have been translated into many languages, and the translated versions have also proven to be reliable and valid. The GMFCS is classified into five subscales according to the functional level of gait and sitting for children with CP. It has been translated into 19 languages, including Korean (http://motorgrowth.canchild.ca). The MACS was designed to classify the objective hand function in the daily lives of 4–18-year-old children with CP (Eliasson et al., 2006). It has five subgroups, with level I indicating maximal functional hand ability and level V indicating that a child cannot independently use his or her hands for meaningful action. The MACS has rapidly spread and is now used around the world, translated into 19 languages, like the GMFCS, despite being developed more recently than the GMFCS (http://www.macs.nu). But these tools are not sufficient to evaluate a child’s degree of skilled movement ability, as they only focus on the end performance result rather than considering the qualitative aspects of an action like the compensatory movement. The TCMS is considered a suitable and useful tool to investigate both the end result and qualitative aspects of movement when performing an action. Although research has been continuously done to develop tools to evaluate the trunk control ability of children with CP, the Seated Postural Control Measure (SPCM) (Fife et al., 1991) has low reliability, and the Spinal Alignment and Range of Motion Measure (SAROMM) (Bartlett & Purdie, 2005) only deals with the postural aspects of sitting. Neither is focused on the static or dynamic trunk control ability in a sitting posture. The Segmental Assessment of Trunk Control (SATCo) (Butler, Saavedra, Sofranac, Jarvis, & Woollacott, 2010) was recently developed and only includes static sitting balance; thus, it is difficult to measure dynamic sitting and reaching balance using the measure. The abovementioned tools are not sufficient for examining both the static and dynamic sitting balance of trunk control ability for functional activity, and this has limited their clinical interpretation. The Trunk Impairment Scale (TIS), developed to evaluate the trunk control ability of adults who Table 2 Inter-rater and intra-rater reliability for the TCMS-K (N = 50). Inter-rater reliability
Static sitting balance Selective movement control Dynamic reaching Total TCMS-K
Intra-rater reliability
ICC
95%CI
ICC
95%CI
.994 .995 .987 .998
.990–.997 .991–.997 .977–.992 .997–.999
.994 .986 .947 .996
.990–.997 .975–.992 .908–.969 .993–.998
ICC, intraclass correlation coefficient; CI, confidence interval; TCMS-K, Korean version of Trunk Control Measurement Scale.
J.-Y. Jeon, W.-S. Shin / Research in Developmental Disabilities 35 (2014) 581–590
585
have had a stroke, can examine both static and dynamic sitting balance. It proved to be adequate in clinical and research settings (Verheyden et al., 2004), but it concentrates on the features of adults who have had a stroke. As such, this tool is also difficult to apply to measuring the trunk control of children with CP. The TCMS based on the TIS can be suitable to assess the trunk control ability of children with CP in a seated position. In addition, it is considered better than other tools because it was constructed to assess the clinical features of impaired trunk control ability common in CP. It deals with both the static and dynamic aspects of trunk control on sitting and has also demonstrated high reliability and validity. However, the TCMS has no non-English versions. That is considered quite meaningful, as this study is the first attempt to investigate the reliability and validity of TCMS in other languages to compare it with the various cultures. The results of this study confirmed the high reliability and validity of the TCMS-K, equal to those of the original version of the TCMS; the inter-rater reliability ICC was .987–.998 and the intra-rater reliability ICC was .947–.996; an ICC value >.90 is considered excellent; .75–.90 is good; and
Contents lists available at ScienceDirect
Research in Developmental Disabilities
Reliability and validity of the Korean version of the Trunk Control Measurement Scale (TCMS-K) for children with cerebral palsy Jun-Young Jeon a, Won-Seob Shin a,b,* a b
Department of Physical Therapy, Graduate School of Daejeon University, Daejeon, Republic of Korea Department of Physical Therapy, College of Natural Science, Daejeon University, Daejeon, Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Article history: Received 3 September 2013 Received in revised form 4 January 2014 Accepted 6 January 2014 Available online 27 January 2014
The Trunk Control Measurement Scale (TCMS) was developed by Lieve Heyrman in 2011 to evaluate the clinical features of impaired trunk control ability in patients with cerebral palsy (CP). This study aimed to demonstrate the reliability and validity of the Korean version of the Trunk Control Measurement Scale (TCMS-K) for children with CP. Fifty children with spastic CP (mean age 9.08 3.75) participated in the study. They were classified using the Gross Motor Function Classification System and the Manual Ability Classification System. The intraclass correlation coefficient (ICC) value of the inter-rater reliability for the TCMS-K was .987–.998, and the intra-rater reliability was .947–.996. The Spearman rank correlation coefficient between the TCMS-K and the Gross Motor Function Measure-B dimension was .860. The results of the study support that the TCMS-K has a high reliability and validity, which is similar to the original version. Thus, the TCMS-K is a suitable evaluation tool to assess the qualitative performance of trunk control and sitting balance for children with CP, and we expect that it will be a very useful tool for clinicians and researchers. ß 2014 Elsevier Ltd. All rights reserved.
Keywords: Cerebral palsy Korean version Trunk Control Measurement Scale Reliability Validity
1. Introduction Cerebral palsy (CP) is a non-progressive and permanent motor and sensory disorder resulting from a permanent defect or lesion in the immature brain. It was increased the incidence of cerebral palsy because of the advances in medical technology and increased use of incubators for baby of premature and low birth weight (Molnar & Alexander, 2006). According to the data of The Korean National Statistical Office (2011), the rate of low birth weight and preterm baby from the total live births was 4.6% and 5.2%, respectively. Hong (2004) reported that 716 children of cerebral palsy caused the low birth weight 57.6%, the premature baby 62.9% in Korea. The high prevalence of cerebral palsy caused by extreme prematurity and low birth weight baby of prior to 32 weeks (Schendel, Schuchat, & Thorsen, 2002). People with CP have significant limitations in activities of daily living due to their lack of motor control ability and balance, the presence of primitive reflexes, spasticity, and voluntary movement difficulties according to the extent of brain impairment (Harbourne, Willett, Kyvelidou, Deffeyes, & Stergiou, 2010). The main problems experienced by children with CP are abnormal posture and movement strategies because of their trunk instability, motor and sensory impairment, joint contracture, muscle shortening, and deformed musculoskeletal system with reduced bone mineral density (Goodman &
* Corresponding author at: Department of Physical Therapy, Daejeon University, 62 Daehak-ro, Dong-gu, Daejeon 300-716, Republic of Korea. Tel.: +82 42 280 2294; fax: +82 42 280 2295. E-mail addresses: [email protected], [email protected] (W.-S. Shin). 0891-4222/$ – see front matter ß 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ridd.2014.01.009
582
J.-Y. Jeon, W.-S. Shin / Research in Developmental Disabilities 35 (2014) 581–590
Fuller, 2008). It has been the primary problem in assessing the status of the children (Brogren, Hadders-Algra, & Forssberg, 1998). Nevertheless, assessment and research on children with CP have focused on upper and lower extremities rather than the trunk, and there is a lack of studies dealing with the examination of trunk control measurement (Damiano, Alter, & Chambers, 2009; Klingels et al., 2010). It has being applied in order to solve these problems of cerebral palsy in Korea that invasive methods include the surgical intervention, the Botox injection, and non-invasive methods include the neurodevelopmental treatment, the Bobath approach, the Vojta approach, proprioceptive neuromuscular facilitation, functional electrical stimulation, biofeedback training, postural control. Generally, trunk control while in a sitting posture develops from two to nine months of age, leading to independent head control for typically developing children (Harbourne & Stergiou, 2003). However, children with neurological damage have a delayed trunk control ability and sitting alignment depending on their degree of impairment (Brogren, Forssberg, & HaddersAlgra, 2001; Liao, Yang, Hsu, Chan, & Wei, 2003). Almost all children with CP have difficulty standing independently and have considerable difficulty with and fear of maintaining a standing position, due to the narrow base of support it provides. Therefore, this posture is not suitable to assess their trunk control ability, but the sitting posture can give them freedom in their daily lives due to the wide base of support and stability it provides. It is the best suitable posture to assess their functional level and balance ability, as it is their main posture for activities of daily living (Brogren et al., 1998). There are some clinical outcome measurements to evaluate the gross motor, function and balance for children of cerebral palsy in Korea that the gross motor function measurement, Gross Motor Function Classification System, pediatric balance scale, etc. But, these tools are limited that evaluate the quality aspect of action and compensation movement occurred during the trunk motion on sitting. The Trunk Control Measurement Scale (TCMS) was developed by Lieve Heyrman in 2011 to assess children with CP regarding their clinical features related to trunk control impairment. It is based on the Trunk Impairment Scale (TIS), the tool used to assess trunk control in adults who have had a stroke (Heyrman et al., 2011). It can be scored the quality and compensation movement of trunk motion. To date, no attempts have been made to develop a Korean version of the TCMS. A Korean version of the scale should have the same meaning and validity as the original version to reduce the confusion with English expressions and cultural differences when translated into Korean (Beaton, Bombardier, Guillemin, & Ferraz, 2000). Using a common questionnaire to objectively assess in another country’s language enables countries and their populations’ unique characteristics to be compared to overcome culture and language differences (Bicer, Yazici, Camdeviren, & Erdogan, 2004). Also, the exchange of information and the development of research can be achieved (Pietrobon, Coeytaux, Carey, Richardson, & DeVellis, 2002). The TCMS should particularly be used to translate the Korean version considering the cultural features of Koreans’ sedentary lifestyle (Bae, Jang, Lee, & Kim, 2012). Thus, the purpose of this study was to determine the reliability and validity of the Korean version of the original TCMS developed in English for children with CP. 2. Materials and methods 2.1. Participants This study was participated 50 children (22 males and 28 females) with spastic CP with mean age of 9.08 years from two rehabilitation clinics in Daejeon city. We included participants who able to sit independent or need to only hand and feet support for at 30 min. They could communicate and understand the instructions for the test. We excluded the child who injected the botulinum toxin or performed orthopedic surgery and selective posterior rhizotomy in the last six months, have the problem of visual and equilibrium system. We also excluded other type of CP (Heyrman et al., 2011). We used the Gross Motor Function Classification System (GMFCS) and Manual Ability Classification System (MACS) to classify the functional level of subjects. Ethical approval was gained by the institutional review board of the Daejeon University. We provided the sufficient explanation about research to children and parents. They were asked to sign a written statement in which they formally consented to the inclusion of the study. Their characteristics are presented in Table 1. 2.2. Measurement tools 2.2.1. Trunk Control Measurement Scale: TCMS The TCMS measures main two components of trunk control on sitting that static and dynamic sitting balance. The dynamic sitting balance is divided into the selective movement control and dynamic reaching. The static sitting balance examines static trunk control during the stable movement of upper and lower limbs on sit posture. The selective movement control of dynamic sitting balance evaluates selected movements of trunk within the base of support in three planes (flexion, extension, lateral flexion and rotation). The dynamic reaching of dynamic sitting balance measures the performance of reaching task during the active trunk movement in three plane while maintaining the balance between the beyond the base of support and the center of gravity. The total scale consist the 15 item, included the five, seven and three items of the each subscales. All items are scored the ordinal scale on two, three or four point, and bilaterally performed if related clinical activities. The total score of the TCMS are from 0 to 58 point, and it have the high score the better performance as GMFCS level is lowered. The reliability for the TCMS was the intraclass correlation coefficient (ICC) .98 for inter-rater, ICC .97 test–retest and the Spearman rank correlation coefficient for construct validity was .6–.87 between TCMS and Gross Motor Function Measure
J.-Y. Jeon, W.-S. Shin / Research in Developmental Disabilities 35 (2014) 581–590
583
Table 1 General characteristics of the subjects (N = 50). Variable Age (years) Sex Male Female Types of cerebral palsy Spastic diplegia Secondary quadriplegia Left hemiplegia Right hemiplegia GMFCS I II III IV V MACS I II III IV V Total
N (%) 9.08 3.75 22 (44%) 28 (56%) 35 6 5 4
(70%) (12%) (10%) (8%)
12 9 10 19 0
(24%) (18%) (20%) (38%) (0%)
15 15 18 2 0 50
(30%) (30%) (36%) (4%) (0%) (100%)
GMFCS, Gross Motor Function Classification System; MACS, Manual Ability Classification System.
(GMFM), excluded A dimension. Thus, the original TCMS has been demonstrated the reliability and construct validity (Heyrman et al., 2013, 2011). 2.2.2. Gross Motor Function Measure: GMFM The GMFM is used to the most common and standardized assessment tool to examine gross motor functional ability for CP. It has been demonstrated the reliability and validity to evaluates the actually possible functional ability and the degree of enhanced movement skill for children (Palisano et al., 2000). It has five dimensions: (A) lying and rolling, (B) sitting, (C) crawling and kneeling, (D) standing, (E) walking, running and jumping. Each item is scored from zero to three, the lower of the GMFM score means the lower of the gross motor function ability. The inter-rater ICC of the GMFM is .89 and the intrarater ICC is .99 (Russell et al., 1989). This study evaluated the only dimension B of the sitting area in order to be compared with the TCMS-K. Dimension A and B could be more easily performed rather than other dimensions for children with CP. Especially, the dimension B was the more relevant with the TCMS than the dimension A in the previous study (Heyrman et al., 2011). Other dimensions (C–E) that included kneeling, standing, walking, running and jumping are considered that limited to compare the balance on sitting because too difficult or impossible to perform for the GMFCS levels II–IV in children with CP. 2.3. Translation procedure The translation procedure of TCMS was consisted of three translation stages included the first and the second translation and the expert committee according to guideline (Beaton et al., 2000). The first translation was achieved using three translators. They were native speakers of Korean, included the two translators who were medical experts and the remaining one who had no medical/clinical background. Each translator translated from original English version to Korean and then made one translation version throughout the discussion. The second translation was achieved by two translators, who were bilingual in English and Korean with no medical knowledge. They translated the result of first stage back into English individually. Finally, the expert committee made a final version of TCMS-K after comparing and analyzing all versions from each stage. The expert committee was consisted of translators participated in the first stage, language professionals, researchers had an experience of questionnaire development and evaluation (Beaton et al., 2000). 2.4. Procedure 2.4.1. Inter-rater and intra-rater reliability Two physical therapists assessed for all subjects with the TCMS-K to examine the inter-rater reliability, the same therapist who conducted the first assessment performed second assessment for all subjects after seven days to investigate the intra-rater reliability.
584
J.-Y. Jeon, W.-S. Shin / Research in Developmental Disabilities 35 (2014) 581–590
2.4.2. Construct validity To determine construct validity of TCMS-K, each therapist examined the GMFM-B dimension and calculated the correlation between the TCMS-K and the GMFM-B dimension (Mawdsley, Hoy, & Erwin, 2000). 2.5. Statistical analysis All data were analyzed using SPSS software version 18.0. Inter-rater and intra-rater reliability for the TCMS-K score was presented by intraclass correlation coefficient (ICC) and the 95% confidence intervals (CI). To investigate construct validity, TCMS-K was compared with GMFM-B dimension score by Spearman rank correlation coefficient. 3. Results 3.1. Inter-rater and intra-rater reliability for the TCMS-K The ICC value of the inter-rater reliability was .987–.998 and the intra-rater reliability was .947–.996. ICC and 95% CI of the subscales and total TCMS-K values are presented in Table 2. 3.2. Construct validity of the TCMS-K The Spearman rank correlation coefficient between the TCMS-K and the GMFM-B dimension was .860 that was significantly positive. 4. Discussion This study was conducted to determine the inter- and the intra-rater reliability and construct validity of the Korean version of the TCMS, the TCMS-K. We introduce a new tool to evaluate the sitting balance and trunk control ability for children with CP. So far, several assessment tools have been developed and used to better understand the examination of motor functions, management protocol decisions, and treatment goal setting for children with CP conducted by many clinicians and researchers (Rosenbaum et al., 2007). For example, the Gross Motor Function Measure (GMFM), Gross Motor Function Classification System (GMFCS), and Manual Ability Classification System (MACS) have demonstrated both high reliability and validity and have been widely used throughout the world (Eliasson et al., 2006; Palisano, Cameron, Rosenbaum, Walter, & Russell, 2006). These tools have been translated into many languages, and the translated versions have also proven to be reliable and valid. The GMFCS is classified into five subscales according to the functional level of gait and sitting for children with CP. It has been translated into 19 languages, including Korean (http://motorgrowth.canchild.ca). The MACS was designed to classify the objective hand function in the daily lives of 4–18-year-old children with CP (Eliasson et al., 2006). It has five subgroups, with level I indicating maximal functional hand ability and level V indicating that a child cannot independently use his or her hands for meaningful action. The MACS has rapidly spread and is now used around the world, translated into 19 languages, like the GMFCS, despite being developed more recently than the GMFCS (http://www.macs.nu). But these tools are not sufficient to evaluate a child’s degree of skilled movement ability, as they only focus on the end performance result rather than considering the qualitative aspects of an action like the compensatory movement. The TCMS is considered a suitable and useful tool to investigate both the end result and qualitative aspects of movement when performing an action. Although research has been continuously done to develop tools to evaluate the trunk control ability of children with CP, the Seated Postural Control Measure (SPCM) (Fife et al., 1991) has low reliability, and the Spinal Alignment and Range of Motion Measure (SAROMM) (Bartlett & Purdie, 2005) only deals with the postural aspects of sitting. Neither is focused on the static or dynamic trunk control ability in a sitting posture. The Segmental Assessment of Trunk Control (SATCo) (Butler, Saavedra, Sofranac, Jarvis, & Woollacott, 2010) was recently developed and only includes static sitting balance; thus, it is difficult to measure dynamic sitting and reaching balance using the measure. The abovementioned tools are not sufficient for examining both the static and dynamic sitting balance of trunk control ability for functional activity, and this has limited their clinical interpretation. The Trunk Impairment Scale (TIS), developed to evaluate the trunk control ability of adults who Table 2 Inter-rater and intra-rater reliability for the TCMS-K (N = 50). Inter-rater reliability
Static sitting balance Selective movement control Dynamic reaching Total TCMS-K
Intra-rater reliability
ICC
95%CI
ICC
95%CI
.994 .995 .987 .998
.990–.997 .991–.997 .977–.992 .997–.999
.994 .986 .947 .996
.990–.997 .975–.992 .908–.969 .993–.998
ICC, intraclass correlation coefficient; CI, confidence interval; TCMS-K, Korean version of Trunk Control Measurement Scale.
J.-Y. Jeon, W.-S. Shin / Research in Developmental Disabilities 35 (2014) 581–590
585
have had a stroke, can examine both static and dynamic sitting balance. It proved to be adequate in clinical and research settings (Verheyden et al., 2004), but it concentrates on the features of adults who have had a stroke. As such, this tool is also difficult to apply to measuring the trunk control of children with CP. The TCMS based on the TIS can be suitable to assess the trunk control ability of children with CP in a seated position. In addition, it is considered better than other tools because it was constructed to assess the clinical features of impaired trunk control ability common in CP. It deals with both the static and dynamic aspects of trunk control on sitting and has also demonstrated high reliability and validity. However, the TCMS has no non-English versions. That is considered quite meaningful, as this study is the first attempt to investigate the reliability and validity of TCMS in other languages to compare it with the various cultures. The results of this study confirmed the high reliability and validity of the TCMS-K, equal to those of the original version of the TCMS; the inter-rater reliability ICC was .987–.998 and the intra-rater reliability ICC was .947–.996; an ICC value >.90 is considered excellent; .75–.90 is good; and
